Recently, digital techniques have been used in many fields such as an audio or a video field, and techniques of a CD player, and R-DAT, a digital VTR, and others have been established.
Among them, a time code control technique has been important together with a basic technique of recording and reproducing a PCM signal, the time code control technique recording time codes in correspondence with the PCM signal and performing sync recording, sync reproducing, and sync editing by connecting a plurality of recording and reproducing apparatuses searching a head of a point of the recording or the reproducing.
In a prior art time code recording and reproducing apparatus, PCM samples are divided into frames, a time code having a minimal unit of a frame number is recorded for each frame, and one time code is given to a plurality of PCM samples belonging to one frame.
Ways of recording time codes which are defined as mentioned above are classified into two types, that is, (1) a rotary head type and (2) a fixed head type.
(1) The rotary head type
The R-DAT records time codes by a rotary head. In the R-DAT, since the frame frequency is generally 2000/60 Hz, time codes of 2000/60 frames are recorded during one second. In this case, since the time codes are recorded and reproduced by the rotary head and thus variations in relative speeds of the head and the tape are small when the tape speed varies, the range of the tape speeds which can be reproduced is large.
In the R-DAT, the reproduction of the time codes is possible between a stationary state and a 200-times speed.
(2) The fixed head type
In general, a rotary head type VTR uses a fixed head in recording time codes. Also in this case, the recording is generally performed in synchronism with a track period recorded by the rotary head. However, in the case where the time codes are recorded by the fixed head, since the recording is possible even when it is out of synchronization with the track recorded by the rotary head, the recording is possible even when time codes having different frame periods are inputted. The collection of preliminary papers of AES Tokyo Convention '87, pages 93 to 94, discloses a way of recording SMPTE time codes into an R-DAT optional track by a fixed head, the time codes having a frame frequency of about 29.97 Hz for a VTR.
(1) Problem in the fixed head type
The time code recording of the fixed head type suits to the transfer of time codes between apparatuses of different types since it can perform the recording independent of the recording frame of the rotary head even when the frame period and the phase of the input time codes disagree with the frame period and the phase of the recording apparatus. However, in the time code recording by the fixed head, since the relative speeds of the head and the tape are determined by the tape speed, there is a problem that the range of the tape speeds which can be reproduced is narrow. In addition, in the case of a recording and reproducing apparatus such as an R-DAT which has a low standard tape speed (8.15 mm/SEC), the reproduction of the time codes at the standard tape speed is difficult. In addition, in the fixed head type, since the accuracy of the attachment location of the fixed head directly affects the accuracy of the recording phase of the time codes, it is difficult to maintain the accuracy. Furthermore, the fixed head type needs the fixed head in addition to the rotary head, and it has a problem that miniaturizing the mechanism is difficult.
Therefore, it is understood that the time code recording by the rotary head type suits to an apparatus such as an R-DAT or a rotary head type VTR which records a main signal by a rotary head.
(2) Problem in the rotary head type
The time code recording of the rotary head type is an excellent way having a wide range of the tape speeds which can be reproduced, and enabling the mechanism to be simplified. However, in the recording way such as time codes of the above-mentioned R-DAT, it is difficult to perform sync editing with an accuracy lower than the frame. This is because times of a plurality of PCM samples in one frame are expressed by one time code and there is no time code below the frame.
FIG. 14 is a timing chart showing the relation between input samples and recording time codes in a prior art time code recording and reproducing apparatus.
In FIG. 14, 10 input samples are inputted as one frame, and an input time code corresponding to the input frame is inputted with an accuracy of a frame unit. A recording frame of a recording apparatus is delayed from the input frame by 0.2 frame. Since the time code has an accuracy of a frame unit only, the recording time code same as the input time code is recorded.
Therefore, the relation between the recorded PCM samples and the time codes is determined uniquely.
In order to accurately determine the relation between the time codes and the PCM samples, it is necessary to provide sync of the frame phase when the time codes are transferred between a plurality of recording and reproducing apparatuses, that is, it is necessary to provide rules that the input frame and the recording frame in FIG. 14 should be made equal. If it is done, there is no phase difference of 0.2 frame in FIG. 14 so that the first PCM sample in a frame agrees with the time denoted by the time code and subsequent PCM samples can be regarded as corresponding to the times being delayed from each other by 0.2 frame intervals.
In the case where the frame sync is required when the time codes are transferred, inconvenience sometimes occurs.
In general, the frame sync of recording and reproducing apparatuses of various types tends to be dependent upon the contents of information handled by the apparatuses or the structures of the apparatuses, and recording and reproducing apparatuses of different types tend to have different frame syncs respectively.
For example, it is about 33.37 msec in agreement with the video frame period in an NTSC color VTR, and it is 30 msec in an R-DAT.
In the case where the time codes are transferred between apparatuses which have different frame periods in this way, frame phase sync such as mentioned above is sometimes not established.
FIG. 15 is a timing chart showing the conversion of continuous time codes having a frame period ratio of 4:3.
In FIG. 15, time codes 1 are time codes transferred from a reproducing apparatus having a frame period of 40 msec, and time codes 2 are time codes converted to be recorded in a recording apparatus having a frame period of 30 msec.
In the case where the time codes 1 are continuous, the time of the first sample of a frame and the time code can be made equal by providing sync between the frame phase of the time codes 2 and the frame phase of the recording apparatus.
FIG. 16 is a timing chart showing the conversion of discontinuous time codes having a frame period ratio of 4:3. In FIG. 16, since the time codes 1 are discontinuous at the point a, the frame phase of the time codes 2 are discontinuous at the point a. In this case, if the frame phase of the time codes 2 and the recording frame phase are newly synchronized at the discontinuous point, the time of the first sample of the frame and the time code can be in agreement. In practice, since the phase of the recording frame can not be changed during the recording operation, sync can not be established and thus there is a problem that the recording must be done with the disagreement between the times of the PCM samples and the time codes.